We used combined measures of local connectivity to map functional alterations in the cerebral cortex in two clinical conditions showing augmented sensitivity to pain as a common feature, but with suspected different origins. The results revealed a distinctive pattern of functional alterations in the cerebral cortex. The most relevant alterations in local connectivity for osteoarthritis patients were identified in the insular cortex, which processes important aspects of the brain response to pain , whereas fibromyalgia patients showed alterations in the sensorimotor cortex extensively affecting the cortical representation of the body. In both groups, cortical dysfunction occurred in the form of lower connectivity, although this may express distinct pathophysiological alterations in osteoarthritis and fibromyalgia.
Sensitized osteoarthritis patients characteristically show increased pain and neural activity in response to mechanical stimulation [1, 2]. However, in the absence of stimulus (i.e., at rest), severe knee osteoarthritis is associated with weakened functional connectivity [8, 9] and reduced gray matter volume in the insular cortex [30–32]. Volume reductions in osteoarthritis may be reversible after successful treatment, indicating that the anatomical alteration is not a predisposing factor but rather the consequence of biological stress from long-lasting pain [33–36]. A valuable MR spectroscopy study demonstrated reversible reductions in N-acetylaspartate (NAA), a marker of neural integrity, that led the authors to conclude that augmented neural activity in pain-processing areas would impair mitochondria-dependent energy supply and reduce resting neuronal activity . In this context, lower local connectivity in the insular cortex in our study is compatible with reduced neural activity during metabolic recovery at rest following repeated activation in osteoarthritis patients.
In fibromyalgia patients, imaging research has demonstrated baseline hypometabolism in the brain , despite patients complaining of spontaneous, resting pain. Therefore, lower local connectivity in our study is again consistent with reduced overall neural activity in the cortex. In fibromyalgia, however, the situation may be interestingly more intricate. Indeed, neurophysiological studies consistently show a deficient inhibition of the cerebral cortex in fibromyalgia patients . A deficient inhibition from gamma-aminobutyric acid (GABA) interneurons may further contribute to local functional connectivity reduction, due to a loss in their synchronization effect . Therefore, lower intra-regional sensorimotor cortex connectivity may express both reduced activity of principal neurons (fewer active neurons) and reduced activity of inhibitory interneurons (reduced synchronization).
In fibromyalgia, therefore, spontaneous pain and bodily discomfort occur without an obvious sensory input increase in individuals with low basal metabolism in a hyperexcitable sensorimotor cortex. This combination of elements is phenomenologically compatible with a state of deafferentation hypersensitivity, where sensitization is proposed to be the effect of gain enhancement of central sensory reception to compensate a “weak” sensory input [40–43]. “Central gain enhancement”, in the form of over-amplification of multi-level auditory signal reception, has been characterized in the case of hearing loss to account for perceptual distortions in the quiet environment (tinnitus) and loudness intolerance (hyperacusis) [40, 44, 45].
The possibility of a relatively weak sensory input exists in more than one sensory domain in fibromyalgia [46, 47], but let us focus on the proprioceptive system. The female sex is an important risk factor for fibromyalgia . Women characteristically show lower muscle (and tendon) mass relative to the total body mass , and thus a less prominent proprioceptive (e.g., muscle spindles and Golgi tendon organs) structure. In addition, a relevant portion (34%) of severe fibromyalgia patients show low levels of insulin-like growth factors , which mediate the growth hormone action of stimulating the collagen synthesis in tendon and skeletal muscle and ultimately optimizing their tensile properties . It has been proposed that a primary imbalance between nociceptive/non-nociceptive input, per se, could favor pain perception without the need of a net increase in nociceptive input . In this context, sensory imbalance might also be accentuated from a net increase in body mass. Although obesity is not a necessary condition, a high body mass index (without a parallel increase in muscle mass) is a significant risk for developing fibromyalgia [48, 52, 53]. Interestingly, increases in body mass index were found to be coupled with expansion of the cerebral cortex representation of the body in children , which may well illustrate the somatosensory imbalance in the cortical reception fields.
Beyond the body composition argument, however, the most compelling evidence for a weak proprioceptive input in fibromyalgia is related to its low efficiency. For example, Ehlers-Danlos and related joint hypermobility syndromes are characterized by both high overlap with fibromyalgia symptoms and dysfunctional proprioception . Also, joint hypermobility is three times more frequent in fibromyalgia than in control groups . And, importantly, upon specific proprioception testing in fibromyalgia cohorts, significant alterations have been demonstrated in muscle contraction [57, 58], muscle relaxation , vibration perception , joint position sense [60–63] and the proprioceptive component of postural control [57, 61, 64–67] with frequent falls [64, 68, 69]. Problems in postural control or balance are also related to altered vestibular and visual dysfunction [64, 69, 70], which may indicate that sensory weakness is not limited to proprioception in fibromyalgia [46, 47].
Therefore, the data support that proprioceptive (non-nociceptive) input may be relatively weak in fibromyalgia. This situation may promote compensatory mechanisms in the form of neural gain enhancement to optimize the signal, albeit at the expense of excessive noise (i.e., pain and sensory discomfort). In the somatosensory system, the phenomenon may be particularly detrimental due to relevant pathway sharing and multi-level crosstalk between mechanical non-nociceptive and nociceptive somatosensory modalities [71, 72], which prevent the amplification of neural signals from being selective. Relevantly, such gain enhancement effects may contribute to low pain tolerance added to the aforementioned primary pain propensity related to nociceptive/non-nociceptive input imbalance itself [18, 71].
Differences in the response to treatments further support distinct pathophysiological mechanisms for pain sensitization in both situations. For example, anti-inflammatory drugs are analgesic in osteoarthritis through the inhibition of nociceptive sensitization [73, 74], whereas opioids, powerful analgesic agents, may paradoxically favor pain sensitization , presumably by promoting the release of inflammatory mediators in the nervous system . In fibromyalgia, conventional analgesic/anti-inflammatory drugs are generally not effective . In contrast, significant effects on fibromyalgia symptom relief have been obtained by stimulating proprioception with exercise or instrumental mechanical stimulation [76, 77].
A limitation in our study is the comparison of clinical populations typically showing significant age differences. We adopted what we considered to be the optimal strategy to circumvent such a limitation. Each clinical condition was compared with a matched control sample and new samples were used to compare groups adjusting for age. In addition, the effect of age was directly tested in the study samples. We found a general effect of age with local connectivity weakening in the sensory cortex. Importantly, the lowest values in connectivity in the sensory cortex were observed in the young group (i.e., fibromyalgia), which would essentially rule out the effect of age in terms of the differences observed between fibromyalgia and osteoarthritis patients.